| { |
| "type": "eo1", |
| "n_obs_steps": 1, |
| "input_features": { |
| "observation.state": { |
| "type": "STATE", |
| "shape": [ |
| 6 |
| ] |
| }, |
| "observation.images.top": { |
| "type": "VISUAL", |
| "shape": [ |
| 3, |
| 720, |
| 1280 |
| ] |
| }, |
| "observation.images.side": { |
| "type": "VISUAL", |
| "shape": [ |
| 3, |
| 720, |
| 1280 |
| ] |
| }, |
| "observation.images.wrist": { |
| "type": "VISUAL", |
| "shape": [ |
| 3, |
| 720, |
| 1280 |
| ] |
| } |
| }, |
| "output_features": { |
| "action": { |
| "type": "ACTION", |
| "shape": [ |
| 6 |
| ] |
| } |
| }, |
| "device": "cuda", |
| "use_amp": false, |
| "use_peft": false, |
| "push_to_hub": true, |
| "repo_id": "autmoate/eo1-put-mug", |
| "private": null, |
| "tags": null, |
| "license": null, |
| "pretrained_path": null, |
| "vlm_base": "Qwen/Qwen2.5-VL-3B-Instruct", |
| "vlm_config": { |
| "transformers_version": "5.5.4", |
| "architectures": [ |
| "Qwen2_5_VLForConditionalGeneration" |
| ], |
| "output_hidden_states": false, |
| "return_dict": true, |
| "dtype": "bfloat16", |
| "chunk_size_feed_forward": 0, |
| "is_encoder_decoder": false, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "problem_type": null, |
| "text_config": { |
| "architectures": null, |
| "output_hidden_states": false, |
| "return_dict": true, |
| "dtype": "bfloat16", |
| "chunk_size_feed_forward": 0, |
| "is_encoder_decoder": false, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "problem_type": null, |
| "vocab_size": 151936, |
| "hidden_size": 2048, |
| "intermediate_size": 11008, |
| "num_hidden_layers": 36, |
| "num_attention_heads": 16, |
| "num_key_value_heads": 2, |
| "hidden_act": "silu", |
| "max_position_embeddings": 128000, |
| "initializer_range": 0.02, |
| "rms_norm_eps": 1e-06, |
| "use_cache": true, |
| "use_sliding_window": false, |
| "sliding_window": null, |
| "max_window_layers": 70, |
| "layer_types": [ |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention" |
| ], |
| "attention_dropout": 0.0, |
| "rope_parameters": { |
| "type": "mrope", |
| "mrope_section": [ |
| 16, |
| 24, |
| 24 |
| ], |
| "rope_theta": 1000000.0, |
| "rope_type": "default" |
| }, |
| "bos_token_id": 151643, |
| "eos_token_id": 151645, |
| "pad_token_id": null, |
| "_name_or_path": "", |
| "model_type": "qwen2_5_vl_text", |
| "output_attentions": false |
| }, |
| "vision_config": { |
| "architectures": null, |
| "output_hidden_states": false, |
| "return_dict": true, |
| "dtype": null, |
| "chunk_size_feed_forward": 0, |
| "is_encoder_decoder": false, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "problem_type": null, |
| "depth": 32, |
| "hidden_size": 1280, |
| "hidden_act": "silu", |
| "intermediate_size": 3420, |
| "num_heads": 16, |
| "in_channels": 3, |
| "patch_size": 14, |
| "spatial_merge_size": 2, |
| "temporal_patch_size": 2, |
| "tokens_per_second": 2, |
| "window_size": 112, |
| "out_hidden_size": 2048, |
| "fullatt_block_indexes": [ |
| 7, |
| 15, |
| 23, |
| 31 |
| ], |
| "initializer_range": 0.02, |
| "_name_or_path": "", |
| "in_chans": 3, |
| "spatial_patch_size": 14, |
| "model_type": "qwen2_5_vl", |
| "output_attentions": false |
| }, |
| "image_token_id": 151655, |
| "video_token_id": 151656, |
| "vision_start_token_id": 151652, |
| "vision_end_token_id": 151653, |
| "tie_word_embeddings": true, |
| "_name_or_path": "", |
| "vision_token_id": 151654, |
| "model_type": "qwen2_5_vl", |
| "output_attentions": false |
| }, |
| "image_min_pixels": 50176, |
| "image_max_pixels": 50176, |
| "use_fast_processor": true, |
| "chunk_size": 8, |
| "n_action_steps": 8, |
| "max_state_dim": 32, |
| "max_action_dim": 32, |
| "num_denoise_steps": 10, |
| "num_action_layers": 2, |
| "action_act": "linear", |
| "time_sampling_beta_alpha": 1.5, |
| "time_sampling_beta_beta": 1.0, |
| "time_sampling_scale": 0.999, |
| "time_sampling_offset": 0.001, |
| "min_period": 0.004, |
| "max_period": 4.0, |
| "supervise_padding_action_dims": true, |
| "supervise_padding_actions": true, |
| "dtype": "bfloat16", |
| "force_fp32_autocast": true, |
| "attn_implementation": "sdpa", |
| "gradient_checkpointing": true, |
| "normalization_mapping": { |
| "VISUAL": "IDENTITY", |
| "STATE": "MEAN_STD", |
| "ACTION": "MEAN_STD" |
| }, |
| "optimizer_lr": 0.0001, |
| "optimizer_betas": [ |
| 0.9, |
| 0.999 |
| ], |
| "optimizer_eps": 1e-08, |
| "optimizer_weight_decay": 0.1, |
| "optimizer_grad_clip_norm": 1.0, |
| "scheduler_warmup_steps": 900, |
| "scheduler_decay_steps": 30000, |
| "scheduler_decay_lr": 0.0 |
| } |
